An overdetermined problem with non-constant boundary condition (Q399399)

An overdetermined problem is obtained by associating to a partial differential equation so many conditions that no solution exists, in general. The situation may be compared, roughly speaking, to a linear algebraic system with more equations than unknowns. Nevertheless, it may well happen that an overdetermined problem does possess a solution in some particular domain \(\Omega_0\): for instance, a famous result by \textit{J. Serrin} [Arch. Ration. Mech. Anal. 43, 304--318 (1971; Zbl 0222.31007)] and \textit{H. F. Weinberger} [Arch. Ration. Mech. Anal. 43, 319--320 (1971; Zbl 0222.31008)], asserts that the following overdetermined problem: \[ \begin{cases} -\Delta u = 1 &\text{in \(\Omega\);} \cr u = 0 &\text{on \(\partial \Omega\);} \cr |\nabla u| = \text{const.} &\text{on \(\partial \Omega\)} \end{cases} \] in a bounded, smooth domain \(\Omega \subset \mathbb R^N\), \(N \geq 2\), is solvable if and only if \(\Omega\) is a ball. The paper under review deals with the more general problem \[ \begin{cases} -\Delta u = 1 &\text{in \(\Omega\);} \cr u = 0 &\text{on \(\partial \Omega\);} \cr |\nabla u(x)| = g(x) &\text{on \(\partial \Omega\)} \end{cases} \leqno{\text{(1)}} \] where \(g \in C^\gamma(\mathbb R^N)\) is a Hölder continuous function satisfying \(g(x) > 0\) for \(x \neq 0\) and such that \(g(tx) = t^\alpha \, g(x)\) for all \(t > 0\) and \(x \in \mathbb R^N\). In the case when \(\alpha\) is a positive real number, different from~\(1\), the authors prove that such a problem admits a solution, i.e., there exists an open set~\(\Omega_0\) (which turns out to be connected) such that (1) is solvable. This is achieved by minimizing the functional \[ J(\Omega) = - {\textstyle\frac12} \, \int_\Omega u_\Omega \, dx, \] where \(u_\Omega\) denotes the solution of \(-\Delta u = 1\) in~\(\Omega\), vanishing on the boundary. The variable~\(\Omega\) is an open set subject to the constraint \[ \int_\Omega g^2(x) \, dx \leq 1. \] The regularity of the boundary~\(\partial \Omega_0\) is also investigated. Moreover, assuming \(\alpha > 1\), the authors prove that the domain~\(\Omega_0\): is starshaped with respect to (and contains) the origin; is monotone (in the sense of set inclusion) with respect to~\(g\); and it is unique (this follows from monotonicity). Convexity of~\(\Omega_0\) is proved in the case \(\alpha \geq 2\) and under the assumption that \(g\) has convex level sets. Further properties of~\(\Omega_0\) are established: radial symmetry when \(g\) is radially symmetric, and symmetry stability when \(g\) is close to be radially symmetric; Steiner symmetry when \(g\) is Steiner symmetric. Finally, the authors prove that \(\Omega_0\) is encapsulated between two prescribed level sets of~\(g\).